Isolation of malignant B cells from patients with chronic lymphocytic leukemia (CLL) for analysis of cell proliferation: validation of a simplified method suitable for multi-center clinical studies

Abstract

Background: Heavy water ((2)H(2)O) labelling of DNA enables the measurement of low-level cell proliferation in vivo, using gas chromatography/pyrolysis isotope ratio mass spectrometry (GC/P/IRMS), but the methodology has been too complex for widespread use. Here, we report a simplified method for measuring proliferation of malignant B cells in patients with chronic lymphocytic leukemia (CLL).

Design and methods: Patients were labelled with (2)H(2)O for 6 weeks; blood samples were obtained at 0, 3, and 6 weeks during (2)H(2)O labelling and 9, 12, and 16 weeks thereafter. Bone marrow was sampled at week 6. Phlebotomy was performed at multiple, non-research clinical sites. CLL cells were isolated in a central laboratory, using a novel RosetteSep-based method; DNA labelling was analyzed by GC/P/IRMS.

Results: In 26 of 29 patients, CLL cell isolation resulted in > or =95% purity for malignant CD5+ B cells; in one patient, malignant cells expressed marginal levels of CD5, and in two others, further sorting of CD5hi malignant cells was required. Cell yields correlated with white blood cell counts and exceeded GC/P/IRMS requirements ( approximately 10(7) cells) >98% of the time; high-quality DNA labelling data were obtained. RosetteSep isolation achieved adequate CLL cell purity from bone marrow in only 64% of samples, but greatly reduced subsequent sort time for impure samples.

Conclusion: This method enables clinical studies of CLL cell proliferation outside of research settings, using a shorter (2)H(2)O intake protocol, a minimal sampling protocol, and centralised sample processing. The CLL cell isolation protocol may also prove useful in other applications. (clinicaltrials.gov identifier: NCT00481858).

Conflict of interest statement

Statement Gregory M Hayes, Robert Busch, Jason Voogt, Marc K Hellerstein, Nicholas Chiorazzi, and Elizabeth J Murphy have either shares or options in KineMed Inc.

Copyright 2009 Elsevier Ltd. All rights reserved.

Figures

Figure 1. Heavy water study protocol

Patients receive 2H2O orally for six weeks, with blood and saliva collected as indicated for determination of body water enrichment. CLL cells were purified from blood samples, and 2H enrichment in DNA was determined. An optional bone marrow biopsy was obtained at week 6.

Figure 2. Purity of CLL cells isolated by the single-step RosetteSep method

(A) Purity was determined as the percentage of CD5+CD19+ cells and plotted for repeated samples obtained from each patient. The bold dotted line indicates the 95% cutoff for sufficient purity. (B) Representative flow cytometry dot plot, showing a > 99% pure population of RosetteSep-isolated CD5+CD19+ B cells isolated at week 0 from patient CLL065. (C) Isotype control used to set cutoffs for positive staining. (D) Plot of forward scatter vs. DRAQ5 staining, showing gating of nucleated cells.

Figure 3. Flow cytometry profiles of isolated CLL cells over 16 weeks of study

RosetteSep™ purified B cells of three representative CLL patients over 16 weeks of study were stained with CD19-FITC and CD5-PE antibodies and analysed by flow cytometry; per cent purity (% double-positive cells) is shown in the upper right quadrant of each plot. (A) CLL065 blood samples consistently provided > 95% pure CLL cells. (B) CLL062 RosetteSep™ purified cells exhibited low CD5 expression, while (C) CLL038 cells consistently presented with an admixture of CD5− (likely polyclonal) B cells.

Figure 4. Cytometic plots of CLL cells purified from peripheral blood and bone marrow via RosetteSep™ method

(A) Plots of samples that contained > 95% pure CD5+/CD19 CLL cells in bone marrow. (B) Plots of samples that required further sorting of bone marrow cells due to the presence on non-B cells or non-clonal B cells, or both. FACS analysis shows similar CD5/CD19 profiles for CLL cells purified from either the bone marrow or peripheral blood.

Figure 5. Yield of RosetteSep™ versus WBC counts

The number of CLL cells isolated per ml of blood via RosetteSep™ procedure is plotted against WBC count for each patient visit. The straight line represents a unit slope, corresponding to 100% recovery of all WBC in the RosetteSep isolate; actual yields are systematically lower, particularly in patients with low WBC counts. R = Pearson correlation coefficient, p

Figure 6. Examples of body water 2H enrichments and CLL cell kinetics in patients

The time course of total body water 2H enrichment (right axis, squares) and % new cells (left axis, diamonds) is shown for two CLL patients who underwent 2H2O intake and sampling as depicted in Fig. 1. Error bars represent SD of n = 4 replicate analyses. (A) Patient with slow CLL cell birth rate. (B) Patient with fast CLL cell birth rate.